Computer systems use memory devices, such as dynamic random access memory (“DRAM”) devices, to store instructions and data that are accessed by a processor. These memory devices are normally used as system memory in a computer system. In a typical computer system, the processor communicates with the system memory through a processor bus and a memory controller. The processor issues a memory request, which includes a memory command, such as a read command, and an address designating the location from which data or instructions are to be read. The memory controller uses the command and address to generate appropriate command signals as well as row and column addresses, which are applied to the system memory. In response to the commands and addresses, data is transferred between the system memory and the processor. The memory controller is often part of a system controller, which also includes bus bridge circuitry for coupling the processor bus to an expansion bus, such as a PCI bus.
Although the operating speed of memory devices has continuously increased, this increase in operating speed has not kept pace with increases in the operating speed of processors. Even slower has been the increase in operating speed of memory controllers coupling processors to memory devices. The relatively slow speed of memory controllers and memory devices limits the data bandwidth between the processor and the memory devices.
In addition to the limited bandwidth between processors and memory devices, the performance of computer systems is also limited by latency problems that increase the time required to read data from system memory devices. More specifically, when a memory device read command is coupled to a system memory device, such as a synchronous DRAM (“SDRAM”) device, the read data are output from the SDRAM device only after a delay of several clock periods. Therefore, although SDRAM devices can synchronously output burst data at a high data rate, the delay in initially providing the data can significantly slow the operating speed of a computer system using such SDRAM devices.
One approach to alleviating the memory latency and bandwidth problem is to use multiple memory devices coupled to the processor through a memory hub. In a memory hub architecture, a system controller or memory hub controller is coupled to several memory modules, each of which includes a memory hub coupled to several memory devices. The memory hub efficiently routes memory requests and responses between the controller and the memory devices. Computer systems employing this architecture can have a higher bandwidth because a processor can access one memory device while another memory device is responding to a prior memory access. For example, the processor can output write data to one of the memory devices in the system while another memory device in the system is preparing to provide read data to the processor. The operating efficiency of computer systems using a memory hub architecture can make it more practical to vastly increase memory capacity in computer systems.
Although there are advantages to utilizing a memory hub for accessing memory devices, the design of the hub memory system, and more generally, computer systems including such a memory hub architecture, becomes increasingly difficult. For example, in many hub based memory systems, the processor is coupled to the memory via a high speed bus or link over which signals, such as command, address, or data signals, are transferred at a very high rate. However, as transfer rates increase, the time for which a signal represents valid information is decreasing. As commonly referenced by those ordinarily skilled in the art, the window or “eye” for the signals decreases at higher transfer rates. With specific reference to data signals, the “data eye” decreases. As understood by one skilled in the art, the data eye for each of the data signals defines the actual duration that each signal is valid after various factors affecting the signal are considered, such as timing skew, voltage and current drive capability, and the like. In the case of timing skew of signals, it often arises from a variety of timing errors such as loading on the lines of the bus and the physical lengths of such lines.
As data eyes of the signals decrease at higher transfer rates, it is possible that one or more of a group of signals provided in parallel will have arrival times such that not all signals are simultaneously valid at a receiving entity, and thus cannot be successfully captured by that entity. For example, where a plurality of signals are provided in parallel over a bus, the data eye of one or more of the particular signals do not overlap with the data eyes of the other signals. In this situation, the signals having non-overlapping data eyes are not valid at the same time as the rest of the signals, and consequently, cannot be successfully captured by the receiving entity.
Clearly, as those ordinarily skilled in the art will recognize, the previously described situation is unacceptable. As it is further recognized by those familiar in the art of high speed digital systems, signal timing and signal integrity are issues that have become increasingly more significant in the design of systems capable of transferring and transmitting information at high speeds because signal characteristics can be affected by many things. As a result, diagnostic analysis and evaluation of signals, whether command, address, or data signals, is becoming a more critical step in the design process for any high-speed digital system. Examples of the types of issues evaluated through diagnostic testing include pattern sensitivity, power and ground sensitivity, voltage margin, signal interactions on a bus, failure analysis, and the like.
The tools typically used in performing diagnostics include logic analyzers, pattern generators, oscilloscopes, and in some cases, modified desktop computers. It will be appreciated that there are many other diagnostic tools that are available, however, one common feature shared by all of these tools is the relatively expensive cost. In many instances, only well funded companies can afford equipment with enough sophistication capable of performing diagnostics on high-speed systems. Often, smaller, less well funded companies must compromise performance of the diagnostic equipment in order to afford the equipment, thus, either making some diagnostic evaluation more difficult, or perhaps, even impossible.
Another issue that often arises with conventional diagnostic tools is the manner in which signals are detected by the diagnostic equipment. More specifically, probes of various sorts are used to couple signals from a signal line for detection by the diagnostic equipment. A problem resulting from this is that the probe can introduce loading effects that change the characteristic of the signal being evaluated. Although probes are specifically designed to have high impedance and low capacitance to minimize loading issues and the introduction of noise, there is still in many cases, an unacceptable level of loading that changes the character of a signal to such a degree that it cannot be accurately evaluated.
Another issue that is specific to performing diagnostics on a memory system is the difficulty associated with obtaining control over the memory bus in order to perform evaluation. The ability to evaluate a memory system often requires that specific signals of interest can be captured and analyzed by obtaining control of the memory bus and monitoring the interaction of the signal with the bus itself. Unless control over the memory bus can be obtained, analysis becomes a difficult task. However, obtaining control over the memory bus is a difficult task in itself because conventional approaches often interfere with the normal operation of the computer system, thus, preventing accurate analysis of the memory system under true, normal operating conditions.
Therefore, there is a need for alternative approaches to performing diagnostic analysis and evaluation on memory modules, including those memory modules for use in a memory hub architecture.
The present invention is directed to a memory hub having an on-board diagnostic engine through which diagnostic testing and evaluation of the memory system can be performed. The memory hub includes a link interface for receiving memory requests for access to memory devices of the memory system and a memory device interface coupled to the memory devices for coupling memory requests to the memory devices for access to at least one of the memory devices. A switch for selectively coupling the link interface and the memory device interface is further included, and a memory hub diagnostic engine is coupled to the switch for coupling control signals to the link interface and the memory device interface to perform diagnostic testing of the memory system. The diagnostic engine includes a maintenance port that provides access to results of the diagnostic testing and through which diagnostic testing commands can be received.